Connellite from Bisbee, Arizona: a SingleCrystal Xray Study David E. Hibbs School of Pharmacy, University of Sydney, NSW 2006, Australia Peter Leverett, Peter A. Williams* School of Natural Sciences, University of Western Sydney, Locked Bag 1797, Penrith South DC NSW 1797, Australia *email: p.williams@uws.edu.au INTRODUCTION We have reported singlecrystal Xray structures of three specimens of buttgenbachite, ca. Cu 36 Cl 6 ( ) 2 (OH) 64 nh 2 O, from the Toughnut mine, Tombstone, Arizona, USA; the Likasi mine, Jadotville, Shaba Province, Democratic Republic of Congo; and the Cole shaft, Bisbee, Arizona (Hibbs et al., 2002, 2003a), together with the structure of a specimen of connellite from the Great Australia mine, Cloncurry, Queensland, Australia (Hibbs et al., 2003b). The nature of the connellitebuttgenbachite solid solution, with connellite being the sulfatedominant, rather than nitratedominant endmember, is complex and the above studies have resolved certain structural difficulties associated with the oxyanions sites in the lattice (Fanfani et al., 1973; McLean and Anthony, 1972). Variability in stoichiometry arises from changes in occupancy of H 2 O, H 3 O + and Cl at the origin of the unit cell, of Cl and at two other sites in the lattice, and the substitution of sulfate for nitrate at one of the latter sites (Table 1). Variable amounts of zeolitic water in structural channels that contain sulfate and nitrate ions serve to further complicate the issue. Connellite is known from a number of mines in Arizona, particularly from around Bisbee (McLean and Anthony, 1972; Anthony et al., 1995). It would have been unlikely that all of the Bisbee material was buttgenbachite and a new singlecrystal structure determination of a specimen from Bisbee, presented below, shows that it is connellite, albeit with some nitrate in the lattice. 1
Members of the series are difficult to distinguish from each other, but it is now certain that both connellite and buttgenbachite occur in the Bisbee deposits. Figure 1. Connellite specimen, 8 cm across, from the Czar mine, Bisbee, Arizona. Harvard Mineralogical Museum collection; photo by Wendell E. Wilson. EXPERIMENTAL The crystal used for the structure determination was from the collection of the Arizona Sonora Desert Museum, Tucson, Arizona (specimen number ASDM 2031). Their specimen is from the Shattuck mine, Bisbee, Arizona, and that it was added to the collection in an early acquisition (Anna Dimitrovic, pers. comm., 2005). A crystal of dimensions 0.2 x 0.3 x 0.4 mm was mounted on an Enraf Nonius CAD4 diffractometer. Unit cell dimensions were determined by leastsquares refinement of the setting angles for 25 reflections within the range 3.73 < θ < 23.23 o and are listed in Table 1 together with associated crystal data. All data were collected at 293(2)K using graphitemonochromatized Mo Kα radiation, corrected for Lorentz and polarization effects and an empirical absorption correction based on phi scans was applied (Walker and Stuart, 1983). The structure was determined by difference Fourier techniques and refined on F 2 by fullmatrix leastsquares methods (Sheldrick, 1997) using all unique data. Scattering factors used were those given by Sheldrick (1997). A summary of crystal data and refinement details is given in Table 2. A starting set of coordinates for copper, chloride and hydroxide ions, which form the Cu(II) polyhedral lattice of all the buttgenbachiteconnellite type structures (Hibbs et al., 2003a, 2003b), including a partially occupied (~50%) chloride ion at the origin, refined smoothly with anisotropic displacement parameters to an R 1 of 0.065. A differencefourier map then revealed the positions of atoms in the large channel parallel to c and centred along (2/3, 1/3, z). As for the Great Australia mine connellite (Hibbs et 2
al., 2003b), the nitrogen, N(1), and oxygen, O(1), atoms of a reduced occupancy (~25%) nitrate ion centered at z = 1/4 and sharing the site with a reduced occupancy (~40%) chloride ion, Cl(4), were found. Further inspection of the difference Fourier map again revealed a small but significant peak ~0.5Å from N(1) along the c axis, which was positioned approximately tetrahedrally at a distance of ~1.5Å to each of the three oxygen atoms of the N(1),O(1) nitrate ion. This is the sulfur atom, S(1), of a sulfate ion of significant occupancy ( ~17%) with three of its oxygen atoms almost coincidental with the three oxygen atoms of the N(1),O(1) nitrate ion. Accordingly, O(1) could be split into O(1a), which was constrained to be 1.20 ± 0.01Å from N(1), and O(1b), which was constrained to be 1.48 ± 0.01Å from S(1) and which gave a nearly tetrahedral setting. The existence of this sulfate ion was confirmed with the subsequent location of the fourth oxygen atom, O(1c), ~1.5Å from S(1) along the c axis, and during subsequent refinement this oxygen atom also was constrained to be 1.48Å from S(1). A subsequent difference Fourier map failed to reveal any further large channel contents centred at 2/3,1/3, z. However, a water molecule, OW, was found to partially occupy a site in the z = 3/4 region (at ~0.5, ~0.25, 3/4). A final refinement was then performed via a sequence of steps in which the occupancies of all the atoms constituting the large channel contents and also the occupancy of the chloride ion, Cl(3), at the origin were refined, resulting in acceptable and similar isotropic thermal parameters for all atoms concerned. Final R 1 and wr 2 values were 0.046 and 0.151, respectively, for 522 reflections with I > 2σ(I). Hydrogen atoms of the hydroxide ions or water molecules were not located in the structure. However, for the requirement of charge balance, a small proportion of the assigned hydroxide ions must be protonated. Final atomic coordinates, site occupancy factors and equivalent isotropic thermal parameters are listed in Table 3. RESULTS AND DISCUSSION The structure of the connellite determined in the present study is essentially the same as that described previously (Hibbs et al., 2003b), where the Cu(II) polyhedral lattice formed by Cu(1), Cu(2), Cu(3), Cu(4), Cl(1), OH(1), OH(2), OH(3) and OH(4) as found for other members of the group (Hibbs et al., 2003a) is again preserved The contents of the large open channels parallel to c wherein the, Cl, and SO 4 2 ions and H 2 O molecules reside, are quite similar, but the sulfate ion content of this particular sample is significantly larger than that of nitrate ion; an empirical formula of Cu 36 Cl 7.86 (SO 4 ) 0.67 ( ) 0.50 (OH) 62.3 5.6H 2 O is found. As before, the nitrate ion is centered on a mirror at (2/3, 1/3,1/4) and shares the site with a chloride ion, the occupancy for which (41%) is slightly greater than that found in the Great Australia mine sample (36%). Again, the present structure shows no further occupation along the z axis at the center of the large channels, but the constriction point, once per c periodicity in these large channels, as defined by water molecules bonded to one of the copper ions, is worthy of comment. In the present structure these lattice water molecules occupy a 6fold site (at x~0.50, y~0.25, z = 3/4) with an occupancy of 16%, and this is similar to the case found in the Likasi sulfatebearing buttgenbachite and various other buttgenbachite crystals (Hibbs et al., 2003a); these had population parameters ranging from 17% to 25%. However, in the Great Australia connellite (Hibbs et al., 2003b) and in the Bisbee 3
buttgenbachite structure (Hibbs et al., 2002) the lattice water molecules are located at a 12fold site (but similarly positioned at x~0.50, y~0.25, z~0.80) with population parameters of 15% and 20%, respectively. Finally, the origin site in the narrow channels running parallel to the c axis and centered around (0,0,z) is again considered to be occupied by a chloride ion, Cl(3), and with a population parameter of 52% this is similar to the 55% occupancy found in the Great Australia connellite (Hibbs et al., 2003b) and to the 60% occupancy found in the Likasi sulfatebearing buttgenbachite (Hibbs et al., 2003a). A comparison of the contents of the large and small channels in the various connellitebuttgenbachitetype structures reported so far is shown in Table 1. Given the complex nature of solid solution in the connellitebuttgenbachite series, definite endmember compositions are impossible to define. Previously (Hibbs et al., 2003b), we have adopted the simple expedient of naming connellite as that phase with mol% SO 4 2 > mol%. In this case, the material we have presently examined from Bisbee is clearly connellite. SUPPLEMENTARY MATERIAL Full lists of crystallographic data excluding structure factor tables have been deposited with the Inorganic Crystal Structure Database (ICSD), Fachinformationszentrum, Karlsruhe, Germany; CRYSDATA@FIZKarlsruhe.DE. Any request to the ICSD for this material should quote the full literature citation and the CSDnumber 415129 (filename BisbeeConnCif). Lists of observed and calculated structure factors are available from the authors upon request. ACKNOWLEDGEMENTS The ARC is acknowledged for financial support. We wish to thank Anna Domitrovic of the Arizona Sonora Desert Museum for supplying the sample from Bisbee. REFERENCES ANTHONY, J.W., WILLIAMS, S.A., BIDEAUX, R.A., and GRANT, R.W. (1995) Mineralogy of Arizona, 3rd Edition, University of Arizona Press, Tucson, pp. 6470, 183. FANFANI, L., NUNZI, A., ZANAZZI, P.F., and ZANZARI, A.R. (1973) The crystal structure of buttgenbachite. Mineralogical Magazine, 39, 264270. HIBBS, D.E., LEVERETT, P., and WILLIAMS, P.A. (2002) Buttgenbachite from Bisbee, Arizona: a singlecrystal Xray study. Neues Jahrbuch für Mineralogie, Monatshefte, 225240. HIBBS, D.E., LEVERETT, P., and WILLIAMS, P.A. (2003a) A single crystal Xray study of a sulphatebearing buttgenbachite, Cu 36 Cl 7.8 ( ) 1.3 (SO 4 ) 0.35 (OH) 62.2.5.2H 2 O, and a reexamination of the crystal chemistry of the buttgenbachiteconnellite series. Mineralogical Magazine, 67, 4760. HIBBS, D.E., LEVERETT, P., and WILLIAMS, P.A. (2003b) Connellitebuttgenbachite from the Great Australia mine, Cloncurry: a crystal structural formula. Australian Journal of Mineralogy, 9, 3942. McLEAN, W.J., and ANTHONY, J.W. (1972) The disordered, zeolitelike structure of connellite. American Mineralogist, 57, 426438. 4
SHELDRICK, G.M. (1997) SHELXL97. A program for the refinement of crystal structures. University of Göttingen, Göttingen, Germany. WALKER, N., and STUART, D. (1983) An empirical method for correcting diffractometer data for absorption effects. Acta Crystallographica, A39, 158166. Table 1. Comparison of the contents of the large and small channels in various connellitebuttgenbachitetype structures, with the % occupancies of the sites shown in parentheses. 1 2 3 4 5 6 (a) Large channels centred along (2/3, 1/3, z) Site #1 Cl Cl Cl Cl Cl Cl z = 1/4 (25%) (45%) (30%) (50%) (36%) (41%) (55%) (18%) (50%) (24%) (25%) z ~ 0.20 2 SO 4 2 SO 4 2 SO 4 and ~ 0.30 (9%) (13%) (17%) Site #2 z ~ 0.08 (37.5%) (24%) and ~ 0.42 H 2 O (6%) Lattice H 2 O at ~ 0.50, ~ 0.25, 3/4 (25%) (17%) (18%) (16%) at ~ 0.50, ~ 0.25, ~0.80 (20%) (15%) (b) Small channels centred along (0, 0, z) Origin H 3 O + Cl Cl Cl Cl Cl z = 0, 1/2 (100%) (50%) (60%) (55%) (55%) (52%) 1. Likasi mine, Jadotville, Shaba Province, Democratic Republic of Congo (Fanfani et al., 1973; Hibbs et al., 2003a). 2. Toughnut mine, Tombstone, Arizona (McLean and Anthony, 1973; Hibbs et al., 2003a). 3. Likasi mine, Jadotville, Shaba Province, Democratic Republic of Congo (Hibbs et al., 2003a). 4. Cole shaft, Bisbee (Hibbs et al., 2002). 5. Great Australia mine, Cloncurry, Queensland, Australia (Hibbs et al., 2003b). 6. Shattuck mine, Bisbee (this work). 5
Table 2. Crystal data and structure refinement details for Bisbee connellite. Empirical formula H 73.50 Cl 7.86 Cu 36 N 0.50 O 72.08 S 0.67 Formula weight 3821.93 Temperature 293(2) K Wavelength 0.71073 Å Crystal system Hexagonal Space group P6 3 /mmc Unit cell dimensions a = 15.7866(15) Å c = 9.1015(12) Å Volume 1964.4(4) Å 3 Z 1 Density (calculated) 3.231 g cm 3 Absorption coefficient 9.905 mm 1 F(000) 1842 Crystal size 0.4 x 0.3 x 0.2 mm θ range for data collection 3.73 to 23.23 Index ranges 8 h 0, 0 k 17, 0 l 10 Reflections collected 522 Independent reflections 522 [R(int) = 0.0000] Refinement method Fullmatrix leastsquares on F 2 Data/restraints/parameters 522/2/63 Goodnessoffit on F 2 1.144 Final R indices [I>2σ(I)]* R 1 = 0.0463, R w = 0.1510 R indices (all data) R 1 = 0.0478, R w = 0.1525 Extinction coefficient 0.0007(4) *The weighting scheme used was w = 1/[σ 2 (F o ) 2 + (0.0938P) 2 + 31.6P] where P = [Max(F o ) 2 + 2(F c ) 2 ]/3 as defined by SHELX97 (Sheldrick, 1997). 6
Table 3. Atomic coordinates (x 10 4 ) and equivalent isotropic displacement parameters (Å 2 x 10 3 ) and site occupancy factors (sof) for Bisbee connellite. U eq is defined as one third of the trace of the orthogonalized U ij tensor. x/a y/b z/c U eq sof Cu(1) 5000 0 0 14(1) 0.25 Cu(2) 2011(1) 0 0 12(1) 0.5 Cu(3) 3364(2) 1682(1) 7500 10(1) 0.25 Cu(4) 3585(1) 165(1) 2500 10(1) 0.5 Cl(3) 0 0 0 11(3) 0.043 Cl(1) 2772(3) 1386(2) 2500 12(1) 0.25 OH1 4504(4) 3707(4) 918(6) 12(1) 1.0 OH2 749(3) 749(3) 989(9) 20(2) 0.5 OH3 6744(4) 7442(4) 1102(6) 11(1) 1.0 OH4 4424(4) 5576(4) 2500 14(3) 0.25 OW 5070(60) 2540(30) 7500 17(16) 0.040 Cl(4) 6667 3333 2500 15(6) 0.034 N(1) 6667 3333 2500 15(6) 0.021 O(1a) 6226(4) 3774(4) 2500 17(5) 0.063 O(1b) 6150(7) 3850(7) 2500 17(5) 0.084 S(1) 6667 3333 2999(54) 14(8) 0.028 O(1c) 6667 3333 4620 14(8) 0.028 7